FR3009311A1 - PROCESS FOR TREATING A USED FRYING OIL. - Google Patents

Abstract

The invention relates to a method of treating a used frying oil, comprising a tangential ultrafiltration step. It also relates to an oil obtained by such a process, and used as a biolubricant, especially for chainsaws.

Description

Process for treating a used frying oil The invention relates to the field of recovery of used cooking oils. It relates more particularly to a process for treating used frying oils for their recovery, in particular as biolubricants, especially for logging chainsaws. The frying process promotes the formation of particles, or polar compounds. WO2004103134 discloses a method of recovering used cooking oil, filtering, and reinjection into the frying.

Also known from US6187355 is a chemical treatment of used cooking oil for reuse in frying. Document FR2795086 still discloses a method of upgrading spent cooking oils as a fuel for the production of electrical energy. Finally a Wikipédia leaflet, entitled "vegetable oil fuel" teaches the difficulty of using used cooking oils as fuel in an engine because these oils migrate to the engine lubricating oil and pollute it by strongly degrading its lubricating qualities. The present invention proposes to go against the above teachings, and proposes a solution that makes it possible to upgrade used cooking oils as lubricants. To this end, the invention relates to a method of upgrading used frying oils, comprising a tangential ultrafiltration step. Such a step makes it possible to eliminate the polar compounds, or at least to reduce them significantly, in particular by passing from a content of such compounds of between 15 and 30% to a content of less than 5%. According to other characteristics: the ultrafiltration cutoff may be less than or equal to 0.8 micron, making it possible to obtain an oil with sufficiently few polar compounds, preferably between 0.4 and 0.8 microns, allowing to obtain satisfactory flow rates during the filtration, the filtration pressure can be between 2 and 4 bars, pressures having shown satisfactory results during the tests, preferably 3 bars, to obtain the best results, the filtration temperature may be between 50 and 60 ° C, temperatures having shown satisfactory results during the tests, preferably 55 ° C, to obtain the best results. On the other hand, the forest is perceived as a sensitive environment. As such, one of the current concerns is to limit the impacts of human interventions. In the field of forestry, lubricants used in machinery or chainsaws are for example sources of pollution. Chain oils are fully dispersed in the environment. At a lower level, the hydraulic circuits of the machines are quite often the object of leaks which can sometimes spread not insignificant quantities of oil. One of the solutions to limit the impact of these oils on the environment is based on the use of so-called "biodegradable" oils, also called "organic oils". This name should not make us forget that we are looking for products with a low impact on the environment, ie having not only a strong biodegradability but also a low ecotoxicity, or even a high level of integration in food chains. . Biodegradability alone is not enough. At the term "organic oil", the term "biolubricant" is therefore preferred. The use of vegetable oils (especially rapeseed) makes it possible to meet these requirements. These oils also have the advantage of being of renewable origin. This is the formulation path that is today preferred for the manufacture of biolubricants. The non-food valorization of agro-resources is an activity that consists in using biomass for various purposes such as the production of biofuels (ethanol, biodiesel ...), biomaterials (bioplastics, composite materials ...) or biomolecules (surfactants, solvents, biolubricants, catalysts ...). Among the sectors of valorization of the biomass, in continuous evolution, one finds the sector of the biolubricants of which the market represents approximately 3% of the lubricant market in Europe. Produced from vegetable materials, biolubricants are used in many industries and have properties that distinguish them from oil-based lubricants. The basic raw material of a biolubricant is one or more oils rich in oleic acid, long-chain fatty acid (18 carbons) more frequently derived from oilseeds. In temperate regions, oilseeds come mainly from sunflower, canola, flax and oleaginous fruits (nuts, hazelnuts, almonds, olives). In the tropics, the main species concerned are coconut palms, oil palms, peanuts and cotton. Other oleaginous species such as castor oil, safflower, sesame, jojoba, crambe and cade may also be used. US5888947 discloses a biodegradable lubricant based on mono-, di-, and triglycerol, with the addition of a vegetable oil and a vegetable fat. Document FR2907790 also proposes a biodegradable lubricant based on glycerol, mixed with mono-propylene glycol. Nevertheless these lubricants are quite expensive, and moreover their prices fluctuate strongly making their availability unreliable. It is therefore an object of the present invention to provide a biolubricant by an inexpensive process, and from resources available in abundance.

For this purpose the invention provides an oil obtained by a method according to the invention. This oil is particular in that it comprises mainly triglycerides, and between 3 and 5% of polar compounds. According to a particular embodiment of the invention, such an oil can be used as a biolubricant, in particular for a chainsaw.

The advantage of the present invention lies in particular in that it makes it possible to provide an outlet for used cooking oils, and in addition it makes it possible to provide an oil as a biolubricant, with good biodegradability, and low ecotoxicity. . Other features and advantages of the invention will emerge from the following detailed description relating to an exemplary embodiment given by way of indication and not limitation. The understanding of this description will be facilitated by reference to the accompanying drawings, in which: FIG. 1 illustrates an exemplary filter element that can be used for the method according to the invention, FIGS. 2 to 5 illustrate the results obtained by a plan. of experiments developed with the Doehlert matrix: - Figure 2 illustrates the response surfaces as a function of the temperature and the pore size at the first hour of filtration, - Figure 3 illustrates the response surfaces according to the pressure and pore size at the second hour of filtration - Figure 4 illustrates the response surfaces as a function of temperature and pore size at the fourth hour of filtration, Figure 5 illustrates the variation of the point of flow over time. FIG. 6 illustrates the variation of the flux as a function of time for two different temperatures; FIG. 7 illustrates the variation of the filtration rate as a function of time in optimal conditions, and with periodic practice of the inverted flow for unclogging ("back flush" " in English). The filtration of the process according to the invention uses for example a tangential filtration element as illustrated in FIG. 1. Such an element may for example have a diameter of 25 mm for a length of 1178 mm. The diameter of the channels can be 3.5mm for about twenty channels. This gives a membrane area of 0.25m2. Such an element can withstand a pressure of 80 bar, a temperature of 150 ° C. and a pH ranging from 0 to 14. The exchange surface can be modulated according to the volume of oil to be treated and can reach tens of hours. m2.

The frying process promotes the formation of particles. The monitoring of the variation of these compounds during the winter and summer collections was analyzed using firstly the official method, chemical method NF EN ISO 8420, and secondly, sensors using of the EBRO brand testers, by the PCT 120 test, from the 3M company, and by lipid class analysis by latroscan thin layer chromatography.

These tests made it possible to observe a similar profile during the two winter and summer collections. The majority of the samples are between 20 and 25% of polar compounds. There is also a dispersion over time. Ceramic membrane ultrafiltration has made it possible to standardize the level of polar compounds. There is a considerable drop in the rate of polar compounds of the order of 76%. The sample of used frying oils before filtration had a content of polar compounds of the order of 18% against approximately 4% after filtration (see Table 1 below).

Table 1 Compounds Method Iatroscan (%) Test EBRO PCT 120 polar Official NF in% (%) ISO 8420 (° / 0) MixHFU before 17,25 ± 1,1 18,32 ± 0,05 18,11 ± 2 , 8 19.4 ± 2.3 MixHFU filtration after 4.28 ± 1.12 3.6 ± 0.02 3.4 ± 1.3 3.9 ± 1.8 filtration MixHFU: mixture of used frying oils This standardization by ultrafiltration has a double advantage. It improves the quality of the oil by stability to weathering reactions, and the acidity level drops significantly from 5.31% to 2.46%. Thanks to this standardization, these frying oils can be used as a biolubricant base. The biolubricant derived from these oils can be used in winter weather conditions of up to -14 ° C without the addition of antifreeze additive or a formulation at lower economic cost for this parameter. In order to optimize filtration parameters such as temperature, pressure, and filtration rate, an experimental design was developed with the Doehlert matrix. The use of the response surface methodology has made it possible to optimize the membrane ultrafiltration process from experiments conducted on a pilot scale (Batch of 300 liters of used frying oils). The influence of the effects and interactions between pressure, temperature and membrane porosity parameters was determined using an experimental design generated using Doehlert's uniform gratings allowing the study of up to 7 levels taking into account experimental constraints. The quadratic models obtained for the flow rate and the content of polar compounds during the filtration make it possible, using the response surfaces and the optimal path analysis, to find the optimal setting of the parameters of the filtration module. The oils obtained can subsequently be formulated to meet the mechanical, ecological and toxicological constraints inherent for example in the forest cutting operation. The measured responses are viscosity, pour point, rate of polar compounds. The applied stresses are as follows: - temperature (7 levels): from 40 ° C to 70 ° C, see figures 2 and 4 - the pressure (5 levels): from 2 bars to 6 bars, see figure 3 - pore size membranes used (3 levels): 0.1 μm, 0.45 μm and 0.8 μm, see FIGS. 2 to 4. The results are shown in FIGS. 2 to 4, which illustrate response surfaces according to pore size and temperature (Fig. 2 and 4) and according to pore size and pressure (Fig. 3). The optimum path is obtained with a pour point of about -14 ° C (see Fig. 5) a pressure of 3 bar, a temperature of 55 ° C, and a membrane pore cut-off of 0.8 pm. This reduces the polar compounds up to 3.65% against about 19% before filtration. These tests also show that there is a strong interaction between the pressure and porosity factors of the membrane, and that the pressure, contrary to what would be expected by those skilled in the art, does not make it possible to increase the flow, but the reduced on the contrary. This is probably due to the fact that an increase in pressure quickly produces clogging of the pores.

The influence of polar compounds on the membrane flux is discussed below (see Figures 6 and 7). These compounds alter the quality of the oil by thermooxidative reactions and are also the cause of progressive clogging of the membrane. Two phenomena can explain this decrease in membrane flux: polarization of concentration and clogging. In this case, a secondary polarization is observed. It is attributed to the formation of a layer that appears when the concentration on the membrane surface reaches the solubility limit of macromolecules, and which acts as an additional membrane thickness and creates additional resistance to membrane fluxes. The second factor affecting flow loss is clogging. It modifies the filtering properties of the membrane in use as a result of the deposition or adsorption of the material (polar compounds). It can be irreversible and depends on the weather. It is due to the adsorption, deposition and accumulation of solutes or particles on the surface and inside the pores. The consequence observed is the decrease in membrane flux which is accompanied by an increase in the retention rate of the membrane.

We are now interested in the influence of the transmembrane pressure on the flow. The flow can be connected to the membrane flow by multiplying it by the surface of the membrane. As part of the test campaign mentioned above, similar tests to that illustrated in FIG. 6 were made at various temperatures, pressures and membrane porosity sizes. Over a period of about 4 hours the results show a decrease in flow regardless of the porosity size of the membrane. For a pressure of 3 bars at a temperature of 50 ° C., the 0.1 μm membrane has a maximum flow rate of 0.35 kg / h, a flow rate attained at the first hour of filtration; this flow rate decreases to 0.28 kg / h at the 4th hour of filtration. For a pressure of 5 bar at 50 ° C, it has a maximum flow rate of 10.24 kg / h, flow achieved at the first hour of filtration; this flow rate decreases to 6.73 kg / h at the 4th hour of filtration. The 0.45 μm membrane has a maximum flow rate, at a pressure of 3 bar at 40 ° C., of 3.35 kg / h (1st hour of filtration), drop to 0.80 kg / h (4th hour of filtration). For a pressure of 5 bar at 40 ° C., the 0.45 μm membrane has a flow rate of 2.06 kg / h (1st hour of filtration) and then drops to 1.82 kg / hour (4th hour of filtration). . With a pressure of 3 bars at 60 ° C., the 0.8 μm porosity membrane has a flow rate of 66.68 kg / h at the first hour of filtration (only 14.25 for 5 bars at 60 ° C.). This flow rate drops at the 4th hour to the value of 7.23 kg / h (7 kg / h for 5 bars at 60 ° C.). The rapid drop in flow observed at each membrane cut-off point during the first few moments of filtration is probably due to a superficial clogging of the membrane. A sharp reduction in the slope is observed, followed by a plateau. For the pressures of 2 and 6 bar it is observed that the plateau is reached after one hour of filtration. At 55 ° C., the recorded flow rates for the 0.45 μm cut-off membrane are of the order of 11.57 kg / h for 2 bars and 0.29 kg / h for 6 bars. In the case of a low pressure (1 or 2 bars), the driving force imposed is not sufficient to obtain a high flow rate. On the other hand, a too high pressure (6 bar) accentuates the clogging phenomenon. In this case the colloidal particles, the unsaturated fatty acids, penetrate more easily into the membrane. The flow then decreases rapidly to reach 0.58 kg / h for the 0.45 μm membrane. The same situation is observed for all other 0.1 μm and 0.8 μm membranes. However, in the case of intermediate pressures (3 and 4 bars), clogging of the pores is slower. After 3 hours of experimentation the flow rates are close to those recorded for a pressure of 6 bars.

The expected response for optimal permeate flow after data processing with the Doehlert matrix is to work at 3 bar. Values between 2 and 4 bar give satisfactory results.

The clogging observed is at the origin of major consequences that will affect the performance of the process. We propose a declogging by inversion of the flow thanks to a manual back flush. We are now interested in the influence of temperature on the flow. The evolution of the filtrate flow as a function of time is shown in FIG. 6 with the tests carried out with the experimental design for the 0.8 μm membrane at a pressure of 3 bars for the temperatures of 45 ° C. and 60 ° C. vs. After 2 hours of filtration, the profile of the two curves is similar and the permeate flow rates, proportional to the membrane flux, are almost identical whatever the temperature. A rise in temperature should result in a decrease in viscosity with consequent increase in flow, the oil having a rheological behavior of Newtonian type. Figure 6 shows instead that the heating of the oil leads to a decline in flow. The recorded flow rates decrease as a function of time, which reflects a clogging phenomenon, which inhibits the expected gain. These results also show that a high rise in temperature does not improve the permeate flux and therefore the performance of the device. In addition, an increase in temperature also affects the quality of the oil which can lead to thermo-oxidative weathering reactions. The expected response for optimum permeate flow after data processing with the Doehlert matrix is to work at 55 ° C. Values between 50 and 60 ° C give satisfactory results. The following is interested in the influence of the polar compounds on the pour point. The tests were conducted with a pilot to implement a unit volume of 50 liters of used cooking oil mixture. The semi-industrial pilot is used, either in "recycling" configuration where the permeate and the retentate are continuously reintroduced into the feed tank in order to work at constant concentration, or in the "production" configuration for which the permeate is then removed. the retentate returns to the feed tray. Figure 5 shows a variation of the crystallization point over time. It is -14.6 ° C at the first filtration and -14 ° C at the 4th hour of filtration. The clogging of the membrane would probably cause this slight drop in the pour point. When clogging occurs, a secondary layer forms on the surface of the membrane. The unsaturated fatty acids can then pass in a very small amount in the permeate, the pilot being used in a permanent recycling configuration.

The responses of the experimental design by the Doehlert matrix show that the optimal parameters are a pressure of 3 bar at 55 ° C. with the 0.8 μm porosity membrane. The experiments carried out were made without using the back flush. The flow rates resulting from these experiments show a gradual decrease in the flow rate over time due to clogging of the membrane. For the validation of the experiment, every 30 minutes a depression was created producing a flow in opposite direction with the back flush to unclog the membrane after measuring the permeate flow. The evolution of the filtrate flow shows an attenuated clogging (see Fig. 7). The measured flow is approximately 133 kg / h / m2, ie a flow rate of 66.5 kg / h. The experiment was repeated 3 times. The oil obtained after filtration according to the invention of a used frying oil mainly comprises triglycerides, and between 3 and 5% of polar compounds. It is thus distinguished from mineral oils or other vegetable oils of the state of the art, which have a much lower content of polar compounds. Nevertheless, the oil obtained by the process according to the invention makes it possible on the one hand to be biodegradable and to have a low ecotoxicity, unlike mineral oils, and on the other hand to be able to be produced at a lower cost, while at the same time ensuring a new outlet for used cooking oils. Although the invention has been described with respect to a particular embodiment, it is understood that it is in no way limited and that various modifications of shapes, materials and combinations thereof can be made. various elements without departing from the scope of the invention.

Claims (7)

REVENDICATIONS1. Process for treating a used frying oil, characterized in that it comprises a tangential ultrafiltration step. 2. Method according to the preceding claim, wherein the ultrafiltration cleavage is less than or equal to 0.8 micron, preferably between 0.4 and 0.8 micron. 3. Method according to one of the preceding claims, wherein the filtration pressure is between 2 and 4 bar, preferably 3 bar. 4. Method according to one of the preceding claims, wherein the filtration temperature is between 50 and 60 ° C, preferably 55 ° C. 5. Oil obtained by a method according to one of the preceding claims, characterized in that it comprises predominantly triglycerides, and between 3 and 5% of polar compounds. 6. Use of an oil according to the preceding claim as a biolubricant. 7. Use according to the preceding claim for a chainsaw.